International Journal of Information Sciences and Techniques (IJIST) Vol.4, No.

3, May 2014
DOI : 10.5121/ijist.2014.4319 155
An Embedded Module as VIRTUAL TONGUE
for Voiceless
1
SANKAR KUMAR S,
2
JENITHA J,
3
NARMADHA I,
4
SUGANYA A
1
Research Scholar,
2,3,4
Final year UG
Department of Electronics and Communication Engineering
Velammal College of Engineering and Technology, Madurai-625009, Tamil Nadu
ABSTRACT
There are several human disabilities in nature of which speech impaired people find difficulty in
communicating with others, which is very important to convey their messages without speech. In this
paper, to make them self reliable and independent, with the advent of embedded systems technology an
embedded handheld icon based assistive device as Virtual Tongue for Voiceless, which speaks for
severely speech disordered people by simply pressing icons appropriately to fulfill their needs, is proposed.
This proposed module comprises a microcontroller based player to play voice messages, Secure Digital
(SD) card reader, Universal Serial Bus (USB) port, icon based remote keypad, audio amplifier and speaker
along with the benefits like portable, reliable, user friendly, affordable cost, low power consumption and of
course speech in regional language with good clarity. The proposed system is designed to produce speech
regardless of time length, audible to the neighbors, based on the request from the user by pressing the icons
thereby this module deserves inarticulate people. An extended version with a feature of converting text into
voice by adding a circuit, with which any text fed through a keyboard can be converted into speech, is also
discussed.
KEYWORDS
Speech impaired, assistive device, voice synthesizer, text to speech
I. INTRODUCTION
In the recent years, there is a rapid increase in the number of speech - disabled victims due to
several reasons like by birth, oral diseases, accidents, etc and need for the Electronic
Assistive Technology (EST) also increases particularly among physically impaired. Speech
impairment is a communication problem in which the normal speech is disrupted due to
articulation problems, because they are rendered helpless and unable to communicate with the
outside world. To facilitate an easy communication for a verbally challenged people and
overcome the hindrances in efficient communication, researchers have been developing ESTs and
many of them are based on sign language, which is the language used by deaf and dump people
for their communication. Sign language based ESTs are successful among the people those who
are well versed with sign language and also less percentage of speech impaired community is
familiar with sign language. Most of speech impaired people are not aware of this language and a
considerable percentage of them are illiterate. In this paper, an embedded assistive aid, which acts
as virtual tongue, developed for the people with communication complexity, to have a successful
communication like one who without any disability through speech by handling the icons
appropriately provided in the developed handheld module is presented.
International Journal of Information Sciences and Techniques (IJIST) Vol.4, No.3, May 2014
156
Idea behind the development of this module is at first to speak for the voiceless people in order to
fulfill their basic needs thereby to make them self reliable and independent. The developed
module produces voice which is audible to the neighbors of the user so that they can react (or)
respond for the voice message. The advantages of this module are user friendly, affordable cost,
portable, low power consumption and of course no skill requirement to operate because of its icon
based keypad. Since the voice message is available from SD card, the most important benefit of
this module is regardless of time any length and any language voice message can be stored and
played. Because the assistive devices which speak for voiceless, developed so far are with limited
time messages and are not having the feature of multilingual. But, in this module, according to the
region of the user just by changing the SD card alone, the same module can be used to have
successful voice communication.
Finally, in the enhanced version which can be utilized by literate but with speech disorder, instead
of using collection of language specific, the user can convert text-to-speech through keyboard
provided with this module, is also discussed.
II. EXISTING METHODS
An Electronic Speaking Glove is designed to facilitate an easy communication for a verbally
challenged people, in which the gestures from the gloves are converted into a synthesized speech
to convey an audible message to others. The glove is internally equipped with multiple flex
sensors that are made up of bend-sensitive resistance elements.
Articulograph sensor AG-500 based system is placed on the lips of the user was developed. The
sensor calculates the distance between the various reference points assigned to the mouth. The
distance is mapped with equivalent SAMPA code which is fed as an input to the voice generating
software and output is obtained in audio form.
A vocal cord vibration switch was developed which uses accelerometer sensors to detect the
vibrations from the vocal cord. The vocal cord vibration due to speech are considered as high
periodic signals and the vibrations produced due to coughs and swallows are considered as low
periodic signals. These low periodic signals are filtered and the high periodic signals are allowed
to activate the switch which can be connected to computers, or communication devices for further
processing.
The digital speech aid, which is an electronic device developed for elimination or reduction of
stuttering, which uses digital signal processing circuitry, to elicit the desired response
characteristics of the feedback signal.
Kathamala, a Voice Output Communication Aid (VOCA) that uses ISD4004 ChipCorder which
provides single-chip record/playback solutions for 8-16 minute messaging along with a
Microcontroller to provide voice production at low cost.
An assistive aid for speech impaired that uses HM2007, a speech processor IC after trained by
individual voice and developed with PIC Microcontroller to playback the stored words based on
the vibration from acoustic plate.
The methods which exist related to this work are with several limitations (especially time limit)
and involve complexity along with high cost. To overcome, it is decided to design and develop a
handheld user friendly embedded module to speak with no time limit for them at an affordable
International Journal of Information Sciences and Techniques (IJIST) Vol.4, No.3, May 2014
157
cost without any complexity in handling the module to make them self reliable, happy and
independent resulting Virtual Tongue.
II. PROPOSED SYSTEM
In this paper, EST is designed and implemented as virtual tongue which will help the people
whom are unable to speak or having difficulties in speaking. A system-on-chip is used which
contains a microcontroller along with MP3 decoder, control program, I
2
C controller for this
application along with a IR remote keypad, SD card reader, USB port and an audio amplifier with
speaker. Block diagram of the proposed system is shown in Figure.1.
Figure.1 Block diagram of the proposed system
The system is equipped with icon based user friendly keypad. Icons will be selected by the user
and based on the users request the pre-programmed controller, will select appropriate voice track
from the SD card (or) USB port and then the same will be played through audio amplifier to
produce audio output.
II.MPLEMENTATION
The idea behind this proposed system is at first tried with APR9600, which is a 60 seconds
recording and playback voice IC. Though it has multi-level analog storage, the total time length
of stored voice should not exceed 60 seconds, thereby more divisions lead to less message length
and the clarity of the audio output found not clear. Then, in order to increase the message length,
studied about ISD4004 ChipCorder which provides 3 volt single-chip record / playback
solutions for 8-16 minute messaging. In both the cases, the difficulties inferred are the limited
time length and requirement of skill to record / play a message.
To overcome the time length limit, i.e., to store voice message regardless of time and high quality
audio output with clarity, it is decided to introduce the MP3 principle in our system.
International Journal of Information Sciences and Techniques (IJIST) Vol.4, No.3, May 2014
158
II.SD MEMORY CARD
The SD card system is a new mass-storage system based on innovations in semiconductor
technology. This SD card is highly integrated flash memory with serial and random access
capability. It is accessible via a dedicated serial interface optimized for fast and reliable data
transmission. This interface allows several cards to be stacked by through connecting their
peripheral contacts. SD card allows the design of inexpensive players and drivers without moving
parts. The block diagram of SD memory card is shown in Figure.2.
Figure.2 Block diagram of SD Memory card
All units these SD card are clocked by an internal clock generator. The interface driver unit
synchronizes the DAT and CMD signals from external CLK to the internal used clock signal. The
card is controlled by the six line SD card interface containing the signals: CMD, CLK,
DAT0~DAT3. After power up by default, the SD card will use only DAT0.
These SD Card interface can operate in two different modes:
. SD Card mode
. SPI mode
Host system can choose either one of modes. SD Card mode allows the 4-bit high performance
data transfer. SPI mode allows easy and common interface for SPI channel.
International Journal of Information Sciences and Techniques (IJIST) Vol.4, No.3, May 2014
159
II.USB
USB is a simple and flexible interface between a system and peripherals. USB allows hot
swapping. The "hot-swapping" means that the devices can be plugged and unplugged without
rebooting the computer or turning off the device. That means, when plugged in, everything
configures automatically and when unplugged, the host will detect its absence and automatically
unload the driver. USB is a serial protocol and physical link with two pair of wires, with which
transmits all data differentially on a single pair of wires. Another pair provides power to
downstream peripherals. The block diagram of USB Device controller and USB connection
diagram are shown in Figure.3 (a) & (b) respectively.
(a)
(b)
Figure.3 (a) Block diagram of USB Device Controller (b) USB Connection
The USB device controller provides the hardware to interface a USB link to a data flow stored in
a double port memory. It requires a 48 MHz reference clock provided by the clock controller.
This clock is used to generate a 12 MHz Full Speed bit clock from the received USB differential
data flow and to transmit data according to full speed USB device tolerance. Clock recovery is
done by a Digital Phase Locked Loop (DPLL) block.
The Serial Interface Engine (SIE) block performs NRZI encoding and decoding, bit stuffing, CRC
generation and checking, and the serial-parallel data conversion. The Universal Function Interface
(UFI) controls the interface between the data flow and the Dual Port RAM, but also the interface
with the C51 core itself.
When a USB peripheral device is first attached to the network, a process called enumeration
process gets started. This is the way by which the host communicates with the device to learn its
identity and to discover which device driver is required. The enumeration starts by sending a reset
signal to the newly connected USB device. In other words, when the host is powered up and a
device connected or removed from the network, the enumeration process is initiated.
International Journal of Information Sciences and Techniques (IJIST) Vol.4, No.3, May 2014
159
II.USB
USB is a simple and flexible interface between a system and peripherals. USB allows hot
swapping. The "hot-swapping" means that the devices can be plugged and unplugged without
rebooting the computer or turning off the device. That means, when plugged in, everything
configures automatically and when unplugged, the host will detect its absence and automatically
unload the driver. USB is a serial protocol and physical link with two pair of wires, with which
transmits all data differentially on a single pair of wires. Another pair provides power to
downstream peripherals. The block diagram of USB Device controller and USB connection
diagram are shown in Figure.3 (a) & (b) respectively.
(a)
(b)
Figure.3 (a) Block diagram of USB Device Controller (b) USB Connection
The USB device controller provides the hardware to interface a USB link to a data flow stored in
a double port memory. It requires a 48 MHz reference clock provided by the clock controller.
This clock is used to generate a 12 MHz Full Speed bit clock from the received USB differential
data flow and to transmit data according to full speed USB device tolerance. Clock recovery is
done by a Digital Phase Locked Loop (DPLL) block.
The Serial Interface Engine (SIE) block performs NRZI encoding and decoding, bit stuffing, CRC
generation and checking, and the serial-parallel data conversion. The Universal Function Interface
(UFI) controls the interface between the data flow and the Dual Port RAM, but also the interface
with the C51 core itself.
When a USB peripheral device is first attached to the network, a process called enumeration
process gets started. This is the way by which the host communicates with the device to learn its
identity and to discover which device driver is required. The enumeration starts by sending a reset
signal to the newly connected USB device. In other words, when the host is powered up and a
device connected or removed from the network, the enumeration process is initiated.
International Journal of Information Sciences and Techniques (IJIST) Vol.4, No.3, May 2014
159
II.USB
USB is a simple and flexible interface between a system and peripherals. USB allows hot
swapping. The "hot-swapping" means that the devices can be plugged and unplugged without
rebooting the computer or turning off the device. That means, when plugged in, everything
configures automatically and when unplugged, the host will detect its absence and automatically
unload the driver. USB is a serial protocol and physical link with two pair of wires, with which
transmits all data differentially on a single pair of wires. Another pair provides power to
downstream peripherals. The block diagram of USB Device controller and USB connection
diagram are shown in Figure.3 (a) & (b) respectively.
(a)
(b)
Figure.3 (a) Block diagram of USB Device Controller (b) USB Connection
The USB device controller provides the hardware to interface a USB link to a data flow stored in
a double port memory. It requires a 48 MHz reference clock provided by the clock controller.
This clock is used to generate a 12 MHz Full Speed bit clock from the received USB differential
data flow and to transmit data according to full speed USB device tolerance. Clock recovery is
done by a Digital Phase Locked Loop (DPLL) block.
The Serial Interface Engine (SIE) block performs NRZI encoding and decoding, bit stuffing, CRC
generation and checking, and the serial-parallel data conversion. The Universal Function Interface
(UFI) controls the interface between the data flow and the Dual Port RAM, but also the interface
with the C51 core itself.
When a USB peripheral device is first attached to the network, a process called enumeration
process gets started. This is the way by which the host communicates with the device to learn its
identity and to discover which device driver is required. The enumeration starts by sending a reset
signal to the newly connected USB device. In other words, when the host is powered up and a
device connected or removed from the network, the enumeration process is initiated.
International Journal of Information Sciences and Techniques (IJIST) Vol.4, No.3, May 2014
160
II.VTV CONTROLLER
The MP3 voice messages stored in removable SD card (or) USB drive are to be played based on
the request from the user appropriately. To accomplish the process, a programmable device is
required to process the users request, which is named as VTV controller, shown in Figure.4. This
controller contains typical 80C31 microcontroller core with a dedicated MP3 decoder with full
audio interface. Built-in 20-bit stereo audio DAC, 440mW mono Audio power amplifier, IR
receiver, Full speed USB controller, PLL for Audio clock and USB clock, I
2
C (Inter Integrated
Circuit) bus controller and standard full duplex UART.
Figure.4 VTV Controller
I
2
C bus provides good support for communication with various slow, on-board peripheral devices
that are accessed intermittently. Most available I
2
C devices operate at speeds up to 400Kbps. I
2
C
is appropriate for interfacing to devices on a single board, and can be stretched across
multiplboards inside a closed system. I
2
C is a two-wire serial bus.
II.AUDIO AMPLIFIER
The proposed module is developed to produce voice for the voiceless. But, the audio
output from the VTV controller is 440mW, which not audible to the neighbor. So, an
additional audio amplifier circuit is introduced with D6283 Audio IC, a 4.6W dual audio
power amplifier, whose output is audible to a neighbor of this modules user. D6283 is
an audio power IC with built-in two channels developed for portable audio players. Also,
it has built-in thermal shut down protection circuit and low popping noise at power ON.
The circuit diagram of 4.6W audio power amplifier is shown in Figure.5.
International Journal of Information Sciences and Techniques (IJIST) Vol.4, No.3, May 2014
161
Figure.5. 4.6 WAudio Power Amplifier
II.IR REMOTE CONTROL
In this module, an icon based keypad, shown in Figure.6 is used as input device to select the
request in order to produce appropriate voice message. Since the VTV controller has built-in IR
receiver, a wireless transmitter is used with input keypad, which is an IR remote control on which
the icons are provided.
Figure.6. Icon based keypad
The remote controller itself called as transmitter. Infrared remote controller use invisible light
pulses below the visible wavelength spectrum. There is a line of sight between the transmitter
(light source) and the receiver (light detector).The operating range of IR remote control from 30
feet to 100 feet and allows the signal to penetrate on walls and glass cabinet enclosures. IR
transmitter and receiver systems are inexpensive and are generally reliable.The carrier frequency
International Journal of Information Sciences and Techniques (IJIST) Vol.4, No.3, May 2014
161
Figure.5. 4.6 WAudio Power Amplifier
II.IR REMOTE CONTROL
In this module, an icon based keypad, shown in Figure.6 is used as input device to select the
request in order to produce appropriate voice message. Since the VTV controller has built-in IR
receiver, a wireless transmitter is used with input keypad, which is an IR remote control on which
the icons are provided.
Figure.6. Icon based keypad
The remote controller itself called as transmitter. Infrared remote controller use invisible light
pulses below the visible wavelength spectrum. There is a line of sight between the transmitter
(light source) and the receiver (light detector).The operating range of IR remote control from 30
feet to 100 feet and allows the signal to penetrate on walls and glass cabinet enclosures. IR
transmitter and receiver systems are inexpensive and are generally reliable.The carrier frequency
International Journal of Information Sciences and Techniques (IJIST) Vol.4, No.3, May 2014
161
Figure.5. 4.6 WAudio Power Amplifier
II.IR REMOTE CONTROL
In this module, an icon based keypad, shown in Figure.6 is used as input device to select the
request in order to produce appropriate voice message. Since the VTV controller has built-in IR
receiver, a wireless transmitter is used with input keypad, which is an IR remote control on which
the icons are provided.
Figure.6. Icon based keypad
The remote controller itself called as transmitter. Infrared remote controller use invisible light
pulses below the visible wavelength spectrum. There is a line of sight between the transmitter
(light source) and the receiver (light detector).The operating range of IR remote control from 30
feet to 100 feet and allows the signal to penetrate on walls and glass cabinet enclosures. IR
transmitter and receiver systems are inexpensive and are generally reliable.The carrier frequency
International Journal of Information Sciences and Techniques (IJIST) Vol.4, No.3, May 2014
162
of such infrared signals is typically in the order of around 38kHz. Pushing a button on a remote
control sets in motion a series of events that causes the controlled device to carry out a command.
II.RESULTS & DISCUSSIONS
The developed system Virtual Tongue for Voiceless is shown in Figure.7.
Figure.7. Virtual Tongue for Voiceless
Since, the audio files are MP3 format, the clarity of the voice is clear and of course regardless of
the message length time.
To store the voice messages in MP3 format, with the help of online open source TEXT-TO-
SPEECH converter, the messages are converted into speech and are stored in SD card /USB
drive. The advantage of this module is the voice messages can be stored in any language with the
help of the above said converter and hence, this becomes a multi-lingual module for voiceless. In
other words, having SD cards with messages in various languages, this module becomes multi-
lingual just by changing the SD card alone.
II.FUTURE WORK
The developed handheld user friendly module is for people with speech disorder, for which no
skill is required to operate and the limitation in this module is, the pre-recorded messages
assigned for the icons on the keypad only played. But, this can be enhanced for skilled & literate
people by adding a text-to-speech IC TTS256 and any text fed through keyboard (may be a string,
numbers etc.) can also be played.
II.CONCLUSION
The proposed system Virtual Tongue for Voiceless is developed with SD card reader, USB port,
4.6W Audio output and a remote control with icon based keypad. User friendly, portable,
affordable cost and low power consumption are the striking features of this module.
REFERENCES
[1] The amazing digital gloves that gives Voice to the Voiceless by Praveen kumar S Havalagi ,Shruthi
Urf Nivedita, International Journal of Advances in Engineering & Technology, 2013
[2] A Review of Literature on Hand Gesture Recognition for Indian Sign Language by Neha V.
Tavari,A. V. Deorankar, Dr. P. N. Chatur ,International Journal of Advance Research in
Computer Science and Management Studies, 2013
International Journal of Information Sciences and Techniques (IJIST) Vol.4, No.3, May 2014
163
[3] Sign Language Recognition System For Deaf And Dumb People by Sakshi Goyal, Ishita Sharma,
Shanu Sharma,Amity University,Noida,Uttar Pradesh,India,2013
[4] Development of an Assistive Aid for Speech Impaired by P.Aparna, P.Mohanapriya, International
Journal of Scientific and Research Publications, Chennai 2012.
[5] SymbolPath: A Continuous Motion Overlay Module for Icon-Based Assistive Communication ,by
Karl Wiegand,Rupal Patel,Northeastern University,USA,2012
[6] Voice for the voiceless by E.Jeevitha, M.R.Meenakshi, V.K.G.Kalaiselvi, Chennai, Computer
Science & Education (ICCSC) , Augest 2011
[7] Electronic Speaking Glove for speechless patients, a tongue to a dumb by B.Ali,S.Munawwar,
B.Nadeem,Sustainable Utilization and Development in Engineering and Technology, 2011
[8] Dynamic Acoustic for Dumb Using Embedded Interface (DADEI) by Srikkanth. G,Computer
Modelling and Simulation, 2010
[9] Translating the sign of dumb person using ARM processor by Brindhu Kumari.A, Communication
Control and Computing Technologies(ICCCCT),2010